I only sleep once every 2.5 to 3 days

My sleep cycle is (roughly) 56-72 hours awake, followed by 4-6 hours of sleep, after which I am fully rested. I’ve had this sleep pattern at least since I was 13.

Did we ever discover whether this person is real? It is still incredibly important.

I’d go through 15-20 big books a week at that age, a habit that continued and all the way till I started working full-time. Right around that time (early to mid '90s), the Internet as we know it came into being. These days, I spend most of my nights studying and learning new things, primarily via the Internet rather than through printed books.

More. I score in the 99.999th percentile on standardized IQ tests, FWIW. I was accelerated by 5 grades in early grade school (jumped from K to 5), and later started college while concurrently enrolled in high school. I majored in math & physics, and ended up teaching a lot of the classes, rather than taking them - as they were not very challenging. I’m a top exec in a large global tech consulting business, where I earn a high six-figure income, without working very hard at it. I meet a lot of exceptional people in this career, and even still, the consensus opinion I get from others is I’m nearly always the smartest guy in the room. I’ve met my match a few times over the years, but it’s a relatively rare event, and always awesome.

Also I’d be happy to complete any clinically relevant assessments, screenings, diagnostic instruments, clinical interviews, or labs (cost permitting) that any clinical professionals or researchers might be interested in conducting or reviewing.

It’s not a choice, it’s just how I’m built. Most people seem to be built for a 16/8 sleep cycle or an 18/6, some outliers have 20/4, and a few freaks can do all nighters on a regular basis. I’ve yet to meet someone else like me that goes 3 days at a time, then sleeps.

I guess I personally have adjusted to find these stories less interesting for a couple of reasons

  1. they never do anything large-scale and intersting (I guess if you take a random person and give them the skills described, that’s probably the outcome you’d anticipate since most people have low ambition)

  2. it seems always the case that you can beat these people just by being normal and obsessed (e.g. goggins). this shouldn’t be true but seems to be true

but now I am thinking again that it’s legitimately an important thing to study

I messaged that person but did not get a reply :(

I’m a top exec in a large global tech consulting business

So likely Accenture or McKinsey?



I think the best way to run an outlier sleep study would actually be to email the people who run Whoop, or an engineer there, or maybe ask one of their investors (or maybe invest yourself or something like that) I bet they’d tell you if they have discovered any insane outliers, energetic people seem more likely to be using a fitness armband too.

e.g. here’s some of their data, from https://3dtxp19t9eb3fmumt31248pw-wpengine.netdna-ssl.com/wp-content/uploads/2019/04/sleep-stage-performance.pdf

It’s pretty annoying that this graph doesn’t separate different athletes (i.e. im not totally sure but it looks like the bar at 4h apparently doesn’t mean ‘this person only sleeps 4h a night’).

Also just generally weird that we don’t see this graph, for sleep-per-person, anywhere – I’ve spent a few hours now googling about outlier sleepers and did not come across it once. Just odd.

Also looks like a great candidate for genetic engineering/embryo selection. Nobody can argue that decreasing sleep is a bad thing.

And just very strange that outlier IQ would correlate so well with outlier sleep. What’s going on here?!?!!

The regression coefficient for sleep duration on weekends was -6.11 (SE = 2.09), indicating an increase of 6.11 points on fluid IQ scores for each hour of shorter sleep duration.

wow, so 2h a night is +36 IQ points if that relationship roughly holds.

So I’m actually more convinced that reddit post is real, since a fraud probably wouldn’t think to make it up.


It’s worth looking into again, the science banana hasn’t actually done a lit review I think. These are notoriously hard experiments to run.

e.g. see https://pubmed.ncbi.nlm.nih.gov/21075239/ from the same author:

So far, the only systematic review on the influence of sleepiness, sleep duration and quality on
children’s daytime performance exclusively
focused on school performance. Moreover, even
though the reported effects in this meta-analysis
were significant, they were rather small (ranging
from r = 0.069 for the relationship between school
performance and sleep duration to 0.133 for the
relationship between school performance and
sleepiness) (Dewald et al., 2010).
Thus, the relationship between sleep duration,
cognitive processes and intellectual ability may be
more complex than common sense and intuitive
beliefs may imply (the longer, the better).
Furthermore, the assumed causality that sleep
duration affects intellectual ability does not necessarily resist a critical analysis and may actually be
misleading. In fact, a recent study in adolescents
demonstrated that not sleep duration per se determined daytime executive functioning, but rather
subjective sleepiness (Anderson et al., 2009).
Moreover, inter-individual differences in sleepiness and neurobehavioural deficits due to sleep
deprivation are characterized by a strong trait component (Van Dongen et al., 2004). Thus, the cognitive and neurobehavioural impairments after
sleep deprivation may not exclusively be of a universal nature but also be modulated by a strong
individual component.

Research must differentiate between aspects related to general underlying traits and those aspects that are characterized by state-dependent fluctuations.